EP2050187B1 - Drive and method - Google Patents

Description

The invention relates to a drive and a method.

From the EP 0 179 356 a method for controlling an induction machine is known. It is according to there FIG. 2 the current and voltage at the motor is detected. In this case, components of the actual value of the flow are formed from a controller structure which corresponds to a modeling of the machine. These are fed to a flux controller whose output influences the switching states of the inverter supplying the machine as manipulated variables.

From the DE 1 941 312 For example, a method of controlling an asynchronous machine using Hall sensors 3 is known.

From the DE 195 32 477 A1 is a method for starting an asynchronous machine is known in which is set to start the torque setpoint zero.

From the DE 199 752 C1 a method is known for connecting an inverter to an asynchronous motor.

From the WO 2005/018086 A As the closest prior art, a method for the controlled impressing of a stator current and a torque setpoint for a converter-fed rotary field machine is known.

From the US 6,359,416 B1 An adaptive predictive synchronous current regulator method is known.

From the publication NESHEVA D ET AL: "BIAS _ENHANCED CONDUCTANCE IN HYDROGENATED AMORPHOUS SILICONE FILMS" Philosophical Magazine B; Vol. 63, No. 5, May 1, 1991, pages 1139-1148, XP 000568125 ISSN: 0141-8637 is an observation of metastability in hydrogenated amorphous silicon known.

From the US 2004/001291 A1 For example, a method for determining a drift condition for angular velocity control in a synchronous motor is known.

The invention is therefore based on the object to improve the control properties of an electric drive.

According to the invention the object is achieved in the drive according to the features specified in claim 1 and in the method according to the features indicated in claim 15.

Essential features of the invention in the drive are that it comprises an electric motor, which is fed by a converter,
the converter comprising a time-discrete control structure which regulates the stator current of the electric motor by means of adjusting the voltage applied to the motor,
whereby the current of the motor is detected discretely,
wherein the control structure comprises a regulator whose actual value is a first current component of the current,
wherein the setpoint specification of the controller is preceded by at least one setpoint limiter.

The advantage here is that the setpoint limiter can be executed in such a way that the setting limit of the converter, that is to say, for example, the voltage setting limit caused by the intermediate circuit voltage, is not exceeded. In addition, it is so executable that there is no significant feedback and thus no tendency to oscillate. In addition, nevertheless, the fastest possible compensation of setpoint changes is made possible and the control characteristics of the electric drive are thus improved. In particular, the control structure is executable such that the actual value is such that it reaches the setpoint applied to the actual controller setpoint after exactly one time step, ie time grid, but this setpoint value can be generated from the output signal of the setpoint limiter.

In an advantageous embodiment, a time delay element is arranged between the setpoint specification of the controller and the output value of the setpoint limitation, in particular for delaying by a single time step. The advantage here is that the controller is a pilot control can be assigned, which essentially corrects the changes in the setpoint. Thus, the controller only needs to compensate for insignificant deviations.

In an advantageous embodiment, the controller is assigned a pilot control, which adds to the output value of the controller a proportional to the time change of the setpoint specification size. The advantage here is that a controlled system with essentially integrating behavior is already well controlled by the pilot control, so that the controller only needs to compensate for deviations and disturbances.

In an advantageous embodiment, the setpoint limitation requires as input at least the non-delayed setpoint, the time-delayed setpoint and the output value of the controller. The advantage here is that the setpoint limitation can be carried out in such a way that the setting limit of the converter is taken into account and also at least substantial portions of the transfer function of the controlled system, including motor.

In an advantageous embodiment, the controller is a dead-beat controller, in particular adapted to the engine arranged in the controlled system. The advantage here is that the actual value at the controller input can reach the applied setpoint in a single time step, ie time grid.

In an advantageous embodiment of the controller is a linear controller, such as P-controller, a PI controller or a PID controller, in particular with pilot control. The advantage here is that simple, inexpensive controller structures can be used.

In an advantageous embodiment, the current components in a coordinate system are determined from the detected current. The advantage here is that an adapted coordinate system can be used, in particular a co-rotating, so that the computational effort is low.

In an advantageous embodiment, the converter comprises a control structure for each current component that is similar to the control structure for the first current component. The advantage here is that overshoot in all power components is avoidable.

In an advantageous embodiment, the feedforward control is designed as a dead-beat control, that is to say that the motor arranged in the controlled system has an inverse behavior, in particular wherein the setting limit of the converter is not taken into account. The advantage here is that the feedforward control is so providable that the feedforward can already disappear in a time step, the essential deviations of the actual size of the setpoint size.

In an advantageous embodiment, the setpoint limiter SB limits the setpoint such that the voltage setting limit is not exceeded, in particular where this happens in such a way that no other disadvantages arise, such as instabilities of the control loop due to additional feedback. The advantage here is that a vibration behavior does not occur and still the setpoint is reached as quickly as possible and without overshoot.

In an advantageous embodiment, the setpoint limiter is provided in such a way that the limitation to the setpoint value only acts if the output variable, that is to say the limited setpoint value I_setpoint, runs in the direction of the unlimited setpoint value. The advantage here is that feedback and thus vibration behavior can be suppressed.

In an advantageous embodiment, the difference between the manipulated variable U and the control limits U_max and U_min of the inverter is determined, is multiplied by a factor of 1 / K and is added to the time-delayed, ie temporally earlier setpoint I_SollZ1, this sum as the maximum or minimum value (max, min) is supplied to a limiter 50 than to the time-delayed setpoint value I set becomes effective limiting value. The advantage here is that the setpoint is reached as quickly as possible and without overshoot.

In an advantageous embodiment, the limiter comprises a delay element which delays its output value by at least one time step, that is to say a time grid, and uses it to determine the size within the limiter. The advantage here is that feedback and vibration behavior can be suppressed.

In an advantageous embodiment, the limiter comprises means for suppressing feedback. The advantage here is that the vibration behavior is reduced.

In an advantageous embodiment, the limiter has, as input values, the maximum and minimum values (max, min) and the time-delayed setpoint value I setpoint,
the smaller value being determined from max and input, which is compared with the time-delayed output value of the limiter and of which the larger value becomes effective as the upper limit of the limiter member acting on the input,
the greater value being determined from min and input, which is compared with the time-delayed output value of the limiter and of which the smaller value becomes effective as the lower limit of the limiter member acting on the input. The advantage here is that the limit values are dependent on the controller output and thus overshoots can be prevented.

wobei n die Zeitschritte nummeriert, (U _max) die Spannungsstellgrenze ist, (u_r,n ) der Wert der Reglerausgangsgröße U ist, K den gleichen Wert hat wie bei der Vorsteuerung V. Von Vorteil ist dabei, dass Parameter der Regelstrecke, wie K, und die Stellgrenze U_max derart berücksichtigbar sind, dass ein Überschwingen unterdrückbar ist.In an advantageous embodiment is
the upper limit of the limiter 50 $i_{sOll\mathrm{Max}.n}=i_{\mathit{should}.n-1}+\frac{1}{K_v}\cdot \left(U_{\mathrm{Max}}-u_{r.n}\right).$

where n numbered the time steps, ( U _{max) is} the voltage setting limit, ( u _{r, n} ) is the value of the controller output variable U, K has the same value as in the feedforward control V. The advantage here is that parameters of the controlled system, such as K , And the setting limit U_max are considered such that an overshoot is suppressible.

Important features of the invention in the method are that it is provided for controlling the motor current in an electric motor fed by an inverter,
wherein the process is carried out in a time-discrete manner and the stator current of the motor is detected time-discretely,
wherein the stator current of the electric motor is controlled,

wherein the actual value of a first current component of the current is regulated to a desired value specification,
the setpoint input being the output of a setpoint limiter,
where the manipulated variable of the regulator is the voltage applied to the motor.

The advantage here is that the setpoint limiter is executable such that it includes parameters that are related to the controlled system and the control limit. Thus, overshoot is preventable.

Further advantages emerge from the subclaims.

• In der Figur 1 ist eine erfindungsgemäße schematische Darstellung der Reglerstruktur gezeigt. Dabei wird an der elektrischen Maschine, beispielhaft ein Elektromotor M, der Stromistwert I_Ist erfasst. Diese Erfassung erfolgt mit der Stromabtastung SA zeitdiskret, also zu jedem Zeitschritt des zeitdiskreten Reglers einmal.

The invention will now be explained in more detail with reference to figures:

• In the FIG. 1 a schematic representation of the controller structure according to the invention is shown. In this case, the actual current value I_Ist is detected on the electric machine, for example an electric motor M. This detection takes place in a time-discrete manner with the current sampling SA, ie once at each time step of the time-discrete regulator.

Since the machine's current is a multi-dimensional size when the machine has a three-phase supply, the current should be represented as a vector.

This vector can be represented in a coordinate system, such as flow coordinate system, rotor coordinate system or the like.

The actual current value I_IST of the figures is to be understood as a current component in this coordinate system.

For multiple power components, a control structure according to the FIG. 1 applicable. In any case, however, in any case for the torque-forming current component as I_IST is a regulator structure FIG. 1 to use.

The setpoint I setpoint and the actual value I_IST are fed to the current controller, which changes the manipulated variable in accordance with the control deviation. In the present case, the voltage U, that is to say the magnitude of the voltage vector of the supply of the motor M, is preferably the said manipulated variable.

It is important in the invention that the setpoint input for the controller SR is supplied to the controller with a time delay. This makes it possible, since at the new time already the new setpoint specification is known that the control structure via a feedforward V temporally synchronous to the new setpoint input influences the output of the controller SR in the desired manner. This then causes according to the physical laws of the controlled system, comprising the motor M, a change of the detected current component actual value I_IST, which is supplied to the controller SR as an actual value. considered So if you have the time-discrete controller structure, it is clear that the actual value and the setpoint specification of the controller SR belong to the same time grid, ie time step.

It is particularly advantageous in the invention that the controlled system can be considered in approximate approximation as integrator, because the motor has an inductance as an important feature in any case. To FIG. 1 the feedforward V is the difference from the time-delayed and temporally instantaneous setpoint input fed, so the change in the setpoint input. Thus, if the proportionality factor Kv is set correctly, then the integrating behavior of the controlled system is directly simulated. The regulator SR then has to compensate only small deviations and disturbances.

Generally speaking, the control structure according to the invention thus has a controller whose setpoint specification is delayed in time, wherein the controller is connected in parallel with a dead-beat precontrol, which is effective in terms of time without delay.

It is also important in the invention that the manipulated variable U is subject to an adjustment limit, since in the motor supplying converter no arbitrarily high voltages can be produced. Nevertheless, an overshoot of the current value in the invention is avoidable and thus reaching the current limit and the associated emergency shutdown and error messages.

The actual controller SR receives only the delayed setpoint specification and compares it with the more recent at least one time step actual value size.

It is important in the invention that the predetermined setpoint I_Soll is processed and only the result of this processing is supplied to the controller. Here, a setpoint limiter SB is used.

This setpoint limiter SB prevents the setting limit of the manipulated variable from being exceeded. In this case, the desired value is limited such that the voltage setting limit SG is not exceeded, and this is done in such a way that no other disadvantages arise, such as instabilities of the control loop due to additional feedback.

In this case, the setpoint limiter SB uses the output variable U of the controller as input variables, the non-delayed actual setpoint value I_Soll and the time-delayed, that is to say earlier in time, setpoint value I_SollZ1.

In this case, the setpoint limiter acts in such a way that the limitation to the setpoint value only acts if the output variable, that is to say the limited setpoint value I_Soll, runs in the direction of the unlimited setpoint value. The output value of the setpoint limiter SB is denoted by I_SollLim

A detailed embodiment of the setpoint limiter shows FIG. 2 , where in FIG. 3 an embodiment of the limiter 50 is shown.

FIG. 2 shows that the difference between the manipulated variable U and the control limits U_max and U_min of the inverter is determined, multiplied by a factor of 1 / K and is added to the time-delayed, ie earlier in time setpoint I_SollZ1. The result is supplied as a maximum or minimum value (max, min) to a limiter 50. In addition, limiter 50 is also supplied with the time-delayed setpoint value I setpoint as input variable input.

wobei n die Zeitschritte nummeriert, (U _max) die Spannungsstellgrenze ist, (u_r,n ) der Wert der Reglerausgangsgröße U ist, K den gleichen Wert hat wie bei der Vorsteuerung V. Für die untere Begrenzung gilt ein analoger Zusammenhang. FIG. 2 takes into account the context $$i_{sOll\mathrm{Max}.\mathrm{}n}=i_{\mathit{should}.\mathrm{}n-1}+\frac{1}{K_v}\cdot \left(U_{\mathrm{Max}}-u_{r.\mathrm{}n}\right).$$

where n numbered the time steps, ( U _max ) is the voltage setting limit, ( u _{r, n} ) is the value of the controller output variable U, K has the same value as in the feedforward control V. An analogous context applies to the lower limit.

The limiter 50 is in FIG. 3 detailed.

The smaller value of input and max is determined. Then, the larger value is determined from this and the time-delayed output value of the limiter 50. This value then takes effect as the upper limit value.

Similarly, the larger value of input and min is determined. Then, the smaller value is determined from this and the time-delayed output value of the limiter 50. This value then takes effect as the lower limit value.

In this case, in the limiter member 61, the input to the upper and lower limit value is effective.

As limiter 50 in FIG. 2 It is also possible to use a known standard limiter member which limits the input to the extreme values max and min. However, in this case, under certain circumstances, a vibration behavior of the control structure may occur due to inherent feedback. Therefore, it is advantageous to the limiter 50 after FIG. 3 perform. However, other vibration-suppressing measures are feasible.

In FIG. 4 an exemplary course of the values of the variables is shown, the setpoint limitation taking effect when the reference variable jumps. The limited setpoint runs exactly so that the actual value reaches this after a sampling raster. As a result, the control deviation of the controller SR remains zero during the entire tracking and the integrator of the regulator SR does not change its value. This prevents the overshooting of the actual value.

In a preferred embodiment according to the invention, the time delay is executed as exactly one time step, that is to say a time grid, of the time-discrete control structure.

The controller is preferably executable as a time-discrete dead-beat controller. In this case, the controller is designed such that it can already compensate for the control deviation in a time step. This succeeds at least when the controller has a transfer function of the Z-transformed form 1 / (z * G (z)), where G is the transfer function of the controlled system.

LIST OF REFERENCE NUMBERS

• I setpoint
• I_Ist actual value
• I_SollLim Output value of the setpoint limiter SB
• U voltage as manipulated variable
• SR current regulator
• SA current sampling
• V pilot control
• R controller
• M engine
• SG voltage limit
• SB setpoint limiter

• 50 limiters
• 61 limiter member

Claims (15)

1. Drive, including an electric motor, which is supplied by
   a converter,
   the converter including a time-discrete closed-loop control structure which regulates the stator current of the electric motor by setting the voltage applied at the motor,
   the current of the motor being acquired in time-discrete fashion,
   the closed-loop control structure including a closed-loop controller having an actual value as a first current component of the current,
   the setpoint input of the closed-loop controller being coupled with at least one upstream setpoint limiter,
   characterised in that
   the difference between the actuating variable U and the actuating limits U_max and U_min of the converter is determined, multiplied by a factor 1/K, and added to the time-delayed, i.e. earlier, setpoint I_setpointZ1, this sum being forwarded as a maximum or a minimum value (max, min) to a limiter (50) as a limit value that acts on the non-delayed I_setpoint.
2. Drive according to claim 1,
   characterised in that
   a time-delay element is arranged between the setpoint input of the closed-loop controller and the output value of the setpoint limiter, in particular for delay by a single time step.
3. Drive according to at least one of the preceding claims,
   characterised in that
   the closed-loop controller is assigned a pilot control which adds to the output value of the closed-loop controller a variable that is proportional to the time change of the setpoint input.
4. Drive according to at least one of the preceding claims,
   characterised in that
   the setpoint limiter requires as input variable at least the non-delayed setpoint, the time-delayed setpoint and the output value of the closed-loop controller.
5. Drive according to at least one of the preceding claims,
   characterised in that
   the closed-loop controller is a dead beat controller, in particular one that is adapted to the motor disposed in the controlled system.
6. Drive according to at least one of the preceding claims,
   characterised in that
   the closed-loop controller is a linear controller, such as a P-controller, a PI-controller or a PID-controller, in particular with pilot control.
7. Drive according to at least one of the preceding claims, characterised in that
   the current components are determined in a coordinate system from the acquired current.
8. Drive according to at least one of the preceding claims,
   characterised in that
   the converter includes for each current component a closed-loop control structure, which is similar to the closed-loop control structure for the first current component.
9. Drive according to at least one of the preceding claims, characterised in that
   the pilot control is implemented in the form of a dead beat control, that is to say, it has an inverse behaviour with respect to the motor disposed in the controlled system, the actuating limit of the converter, in particular, not being taken into account in this context.
10. Drive according to at least one of the preceding claims,
    characterised in that
    the setpoint limiter SB limits the setpoint such that the voltage actuating limit is not exceeded, this being carried out in particular such that no other disadvantages arise, such as instabilities of the closed-loop control circuit as a result of additional feedback,
    and/or in that
    the setpoint limiter is provided such that the restriction to the setpoint is effective only if the output variable, i.e., limited setpoint I_setpoint, runs in the direction of the unlimited setpoint.
11. Drive according to at least one of the preceding claims,
    characterised in that
    the limiter (50) includes a delay element, which delays its output value by at least one time step, i.e., one time raster, and uses it for variable determination within the limiter (50).
12. Drive according to at least one of the preceding claims,
    characterised in that
    the limiter (50) includes means for suppressing feedback.
13. Drive according to at least one of the preceding claims,
    characterised in that
    the limiter (50) has as input values the maximum and minimum values (max, min) as well as the non-delayed setpoint I_setpoint,
    the smaller value being determined from max and input, compared to the delayed output value of the limiter (50), and the larger value thereof becoming effective as upper limit of the limit element (61) acting on the input,
    the larger value being determined from min and input, compared to the delayed output value of the limiter (50), and the smaller value thereof becoming effective as lower limit of the limit element (61) acting on the input.
14. Drive according to at least one of the preceding claims,
    characterised in that
    the upper limit of the limiter (50) amounts to $i_{setpoint\mathrm{}\mathit{max},n}=i_{\mathit{setpoint},n-1}+\frac{1}{K_v}\cdot \left(u_{\mathit{max}}-u_{r,n}\right),$

    

    n numbering the time steps (U_max ) being the voltage actuating limit, (u_r,n ) being the value of the output variable U of the closed-loop controller, K having the same value as in the pilot control V.
15. Method for regulating the motor current in an electric motor supplied by a converter,
    the method being implemented in time-discrete fashion and the stator current of the motor being acquired in time-discrete fashion,
    the stator current of the electric motor being regulated,
    the actual value of a first current component of the current being regulated towards a setpoint input,
    the setpoint input being an output of a setpoint limiter,
    the actuating variable of the closed-loop controller being the voltage applied at the motor,
    characterised in that
    the difference between the actuating variable U and the actuating limits U_max and U_min of the converter is determined, multiplied by a factor 1/K, and added to the time-delayed, i.e. earlier, setpoint I_setpointZ1, this sum being forwarded as a maximum or a minimum value (max, min) to a limiter (50) as a limit value that acts on the non-delayed I_setpoint.

Images